Re: (fwd) [avr-gcc-list] RFC: ATmega256x and 16-bit function pointers

[Marek Michalkiewicz]

On Sat, Mar 19, 2005 at 07:19:06PM +0100, address@hidden wrote:

> You tackle a question I've encountered when I first say the
> -mcall-prologues code. Why is the prologue sequence not jumped to via
> CALL? It would be somewhat shorter to use CALL, then pop the return
> address off the stack at the beginning of the called prologue. And it
> solves your problem, just a pop-to-eind has to be added to the
> prologue.

The problem is that the jump can be to any of the PUSH instructions
at the beginning of __prologue_saves__ - not necessarily the first one
(if fewer registers need to be pushed).  With CALL, you would have to
pop the return address before pushing anything.

So, -mcall-prologues will probably not be supported at all on the
ATmega256x - its goal was to reduce code size (at the cost of speed)
which was most important on the small chips.  I still need to figure
out how to get the mangled function name, to fix -mcall-prologues in
C++ and to implement the function entry points in low 64K words.

Marek